Various techniques for suppressing a traffic volume of vehicles including ordinary automobiles, large-size automobiles (buses, tracks, and the like), and the like, and preventing traffic congestion have been developed.
For example, PTL 1 discloses one example of such techniques for preventing traffic congestion. A traveling support device described in PTL 1 determines a state of a traffic flow from an acceleration rate of a vehicle and an inter-vehicle distance from another vehicle, switches traveling control of the vehicle depending on the determination result, and thereby prevents transition of the traffic flow to a traffic congestion state.
Further, PTL 2 discloses another technique for preventing traffic congestion. A traffic control system described in PTL 2 detects or predicts a traffic volume around a road section where a traffic volume is locally concentrated and traffic congestion is likely to occur, such as a lane junction location. Then, the traffic control system controls traveling of a vehicle in such a way that a traffic flow of the road section does not make transition to a traffic congestion state.
As a related technique, PTL 3 discloses a technique for calculating a traffic congestion degree, based on reservation information of road use, and controlling the traffic congestion degree by setting a usage toll depending on the traffic congestion degree. Further, PTL 4 discloses a technique for determining whether road section is in a traffic congestion state, based on vehicle position information, and setting a usage toll of the road section depending on the determination result. NPL 1 and NPL 2 disclose a heterogeneous mixture learning technique for generating a prediction model for each group having the same pattern or regularity of data.